Workshop
Set working directory
Visualizations
library(tidyverse)## ── Attaching packages ────────────── tidyverse 1.2.1 ──## ✔ ggplot2 3.0.0 ✔ purrr 0.2.5
## ✔ tibble 1.4.2 ✔ dplyr 0.7.6
## ✔ tidyr 0.8.1 ✔ stringr 1.3.1
## ✔ readr 1.1.1 ✔ forcats 0.3.0## ── Conflicts ───────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()library(ggthemes)
plant_data <- read.csv('plant_data.csv', header = TRUE)
ggplot(plant_data,
aes(x=Biostimulant,
y=Leaves/1000))+ #,fill=TimePoint)) +
facet_grid(.~TimePoint) +
geom_violin() +
xlab('Biostimulants') +
ylab('Leaves dry weight (g)') + theme_igray() +
theme(axis.text.x = element_text(angle=90, vjust=0.5)) +
scale_fill_pander()## Warning: Removed 1 rows containing non-finite values (stat_ydensity).
Data Cleaning
##Check NANs
colSums(is.na(plant_data))## X.SampleID TimePoint Biostimulant Rep Area
## 0 0 0 0 1
## SPAD.avg Leaves Stems Root Tot
## 0 1 0 0 0plant_data[!complete.cases(plant_data),]Imputing missing data
##Imputing NANs with averages of subsets of the data (based on timepoint and biostimulants, thus taking the mean from replicates of the same missing sample)
plant_data$Area <- ifelse(is.na(plant_data$Area),
ave(plant_data[(plant_data$TimePoint==plant_data[which(is.na(plant_data$Area)),'TimePoint'] & plant_data$Biostimulant==plant_data[which(is.na(plant_data$Area)),'Biostimulant']),'Area'],
FUN = function(x) mean(x, na.rm = TRUE)),plant_data$Area)
plant_data$Leaves <- ifelse(is.na(plant_data$Leaves),
ave(plant_data[(plant_data$TimePoint==plant_data[which(is.na(plant_data$Leaves)),'TimePoint'] & plant_data$Biostimulant==plant_data[which(is.na(plant_data$Leaves)),'Biostimulant']),'Leaves'],
FUN = function(x) mean(x, na.rm = TRUE)),plant_data$Leaves)
##Check NANs
colSums(is.na(plant_data))## X.SampleID TimePoint Biostimulant Rep Area
## 0 0 0 0 0
## SPAD.avg Leaves Stems Root Tot
## 0 0 0 0 0Feature engineering
Combine categorical variables
lda_variables <- plant_data %>%
mutate(Biostimulant_TimePoint=paste(Biostimulant, TimePoint, sep = '_'))LDA Area vs TimePoint
library(MASS)##
## Attaching package: 'MASS'## The following object is masked from 'package:dplyr':
##
## selectarea_vs_treatment_set = dplyr::select(plant_data, Area, TimePoint)
# Feature Scaling
area_vs_treatment_set[1] = scale(area_vs_treatment_set[1])
# Applying LDA
lda_area_vs_treatment = lda(formula = TimePoint ~ Area,
data = area_vs_treatment_set)
area_vs_treatment_results = as.data.frame(predict(lda_area_vs_treatment, area_vs_treatment_set))
area_vs_treatment_results = area_vs_treatment_results[c(6, 1)]
names(area_vs_treatment_results) <- c('LDA1_area_vs_treatment')
# Check how much variance the LDA covers
lda_area_vs_treatment$svd^2/sum(lda_area_vs_treatment$svd^2)## [1] 1lda_variables <- cbind(lda_variables, area_vs_treatment_results[1])LDA Biomass vs Timepoint
biomass_vs_treatment_set = dplyr::select(plant_data, Leaves, Root, Stems, TimePoint)
# Feature Scaling
biomass_vs_treatment_set[-4] = scale(biomass_vs_treatment_set[-4])
# Applying LDA
lda_biomass_vs_treatment = lda(formula = TimePoint ~ Leaves+Root+Stems,
data = biomass_vs_treatment_set)
biomass_vs_treatment_results = as.data.frame(predict(lda_biomass_vs_treatment, biomass_vs_treatment_set))
biomass_vs_treatment_results = biomass_vs_treatment_results[c(6,7,8,1)]
names(biomass_vs_treatment_results) <- c('LDA1_biomass_vs_treatment')
lda_biomass_vs_treatment$svd^2/sum(lda_biomass_vs_treatment$svd^2)## [1] 0.972823981 0.025374543 0.001801476lda_variables <- cbind(lda_variables, biomass_vs_treatment_results[1])Statistical exploratory analysis
Correlation plots
library(corrplot)## corrplot 0.84 loadeddummyvars <- model.matrix(X.SampleID~Biostimulant+Biostimulant_TimePoint,lda_variables)
encoded_vars <- cbind(dummyvars[,-1], lda_variables[,which(names(lda_variables)!='LinkerPrimerSequence' &
names(lda_variables)!='BarcodeSequence' &
names(lda_variables)!='Description' &
names(lda_variables)!='InputFileName' &
names(lda_variables)!='X.SampleID' &
names(lda_variables)!='Rep' &
names(lda_variables)!='Biostimulant' &
names(lda_variables)!='TimePoint' &
names(lda_variables)!='Biostimulant_TimePoint')])
corr.mat = cor(encoded_vars, use='complete.obs')
p.mat = cor.mtest(encoded_vars)$p
row.names(p.mat) <- row.names(corr.mat)
colnames(p.mat) <- colnames(corr.mat)
corrplot(corr.mat, type='lower',
tl.pos = 'lt',
addCoefasPercent = TRUE,
method = 'square',
number.cex = .4, tl.cex = 0.8,
order = 'FPC', tl.col="black",
p.mat = p.mat, sig.level = 0.01,
insig = 'label_sig',
pch.cex = .3,
pch = '*') +
corrplot(p.mat, type = 'upper', add = T, tl.pos = 'n', method = 'circle',
col =rainbow(40, start = 5/6, end = 4/6), cl.lim =c(0,1), diag = FALSE)
## Leaves Tot Root
## Leaves 1.00000000 0.995723837 0.96940097
## Tot 0.99572384 1.000000000 0.99375599
## Root 0.96940097 0.993755990 1.00000000
## Stems 0.98362181 1.002198386 0.98875469
## LDA1_biomass_vs_treatment 1.16468160 1.193950825 1.21145877
## Area 0.93359382 0.864161010 0.81455220
## LDA1_area_vs_treatment 0.93359382 0.864161010 0.81455220
## Biostimulant_TimePointSumagrow_T3 0.34340703 0.586056253 0.60882466
## Biostimulant_TimePointControl_T3 0.39367886 0.611397006 0.59845048
## Biostimulant_TimePointInocucor_T3 0.35351495 0.580923657 0.58575973
## Biostimulant_TimePointPathway_T3 0.34334009 0.585020527 0.58565901
## Biostimulant_TimePointEndomaxx_T3 0.51651659 0.324622323 0.58424893
## SPAD.avg 0.43433526 0.155594941 0.35266336
## Biostimulant_TimePointSumagrow_T2 0.24553066 -0.003210153 0.19821858
## Biostimulant_TimePointPathway_T2 0.24093189 0.006330031 0.21500864
## BiostimulantSumagrow 0.22194174 0.225156785 0.02630018
## Biostimulant_TimePointInocucor_T2 0.25237861 0.200961960 -0.06869508
## Biostimulant_TimePointControl_T2 0.22641263 0.178532746 -0.08003617
## BiostimulantPathway 0.21122233 0.225802627 0.02099342
## BiostimulantInocucor 0.23159250 0.218743512 0.19786498
## Biostimulant_TimePointEndomaxx_T2 0.19426812 0.182758507 0.16405429
## BiostimulantEndomaxx 0.16011999 0.175250132 0.20207966
## Biostimulant_TimePointInocucor_T1 0.09170286 0.078345856 0.08324646
## Biostimulant_TimePointSumagrow_T1 0.69918270 0.625862955 0.67662286
## Biostimulant_TimePointPathway_T1 -0.06037137 0.450450390 0.30431481
## Biostimulant_TimePointControl_T1 0.44454594 0.686100993 0.84565505
## Biostimulant_TimePointEndomaxx_T1 0.53233412 0.787571033 0.68139120
## Biostimulant_TimePointPathway_T0 0.67526442 0.665134628 0.61804883
## Biostimulant_TimePointSumagrow_T0 0.45171076 0.732569343 0.68105724
## Biostimulant_TimePointInocucor_T0 0.58350393 0.550206651 0.60703526
## Biostimulant_TimePointEndomaxx_T0 0.72625469 0.624205873 0.59059149
## Stems LDA1_biomass_vs_treatment
## Leaves 0.983621806 1.16468160
## Tot 1.002198386 1.19395082
## Root 0.988754694 1.21145877
## Stems 1.000000000 1.19011866
## LDA1_biomass_vs_treatment 1.190118657 1.00000000
## Area 0.823698571 1.14925044
## LDA1_area_vs_treatment 0.823698571 1.14925044
## Biostimulant_TimePointSumagrow_T3 0.591388340 0.97002406
## Biostimulant_TimePointControl_T3 0.615330622 0.95106366
## Biostimulant_TimePointInocucor_T3 0.583724695 0.94233301
## Biostimulant_TimePointPathway_T3 0.601218946 0.94390155
## Biostimulant_TimePointEndomaxx_T3 0.522098423 0.94730194
## SPAD.avg 0.320624190 0.69525414
## Biostimulant_TimePointSumagrow_T2 0.185812419 0.54702235
## Biostimulant_TimePointPathway_T2 0.202540305 0.56688202
## BiostimulantSumagrow 0.220367806 0.59647682
## Biostimulant_TimePointInocucor_T2 0.190674869 0.48410639
## Biostimulant_TimePointControl_T2 0.165622680 0.47486454
## BiostimulantPathway 0.233099163 0.59218386
## BiostimulantInocucor 0.007143058 0.54902220
## Biostimulant_TimePointEndomaxx_T2 -0.028229295 0.51591908
## BiostimulantEndomaxx -0.038512433 0.56393470
## Biostimulant_TimePointInocucor_T1 -0.144610985 0.43846561
## Biostimulant_TimePointSumagrow_T1 0.556747101 0.77190212
## Biostimulant_TimePointPathway_T1 -0.022550517 -0.10627022
## Biostimulant_TimePointControl_T1 0.755480590 0.01715083
## Biostimulant_TimePointEndomaxx_T1 0.770097755 -0.05805714
## Biostimulant_TimePointPathway_T0 0.583524966 -0.08345161
## Biostimulant_TimePointSumagrow_T0 0.696905277 -0.08708204
## Biostimulant_TimePointInocucor_T0 0.510865654 0.70979950
## Biostimulant_TimePointEndomaxx_T0 0.565031908 -0.08952663
## Area LDA1_area_vs_treatment
## Leaves 0.93359382 0.93359382
## Tot 0.86416101 0.86416101
## Root 0.81455220 0.81455220
## Stems 0.82369857 0.82369857
## LDA1_biomass_vs_treatment 1.14925044 1.14925044
## Area 1.00000000 1.56805979
## LDA1_area_vs_treatment 1.56805979 1.00000000
## Biostimulant_TimePointSumagrow_T3 0.78449689 0.78449689
## Biostimulant_TimePointControl_T3 0.68032365 0.68032365
## Biostimulant_TimePointInocucor_T3 0.71707683 0.71707683
## Biostimulant_TimePointPathway_T3 0.71058866 0.71058866
## Biostimulant_TimePointEndomaxx_T3 0.76352779 0.76352779
## SPAD.avg 1.11187992 1.11187992
## Biostimulant_TimePointSumagrow_T2 0.72931728 0.72931728
## Biostimulant_TimePointPathway_T2 0.71262026 0.71262026
## BiostimulantSumagrow 0.60320515 0.60320515
## Biostimulant_TimePointInocucor_T2 0.72457336 0.72457336
## Biostimulant_TimePointControl_T2 0.73897053 0.73897053
## BiostimulantPathway 0.54579396 0.54579396
## BiostimulantInocucor 0.59601466 0.59601466
## Biostimulant_TimePointEndomaxx_T2 0.69574722 0.69574722
## BiostimulantEndomaxx 0.54720190 0.54720190
## Biostimulant_TimePointInocucor_T1 0.63552361 0.63552361
## Biostimulant_TimePointSumagrow_T1 0.08381664 0.24399966
## Biostimulant_TimePointPathway_T1 0.13260546 0.30249675
## Biostimulant_TimePointControl_T1 0.88853235 0.01096427
## Biostimulant_TimePointEndomaxx_T1 0.58601563 -0.02740427
## Biostimulant_TimePointPathway_T0 0.04483800 -0.34001125
## Biostimulant_TimePointSumagrow_T0 0.38003179 -0.33501550
## Biostimulant_TimePointInocucor_T0 -0.25970187 -0.09951884
## Biostimulant_TimePointEndomaxx_T0 -0.02269113 -0.33401142
## Biostimulant_TimePointSumagrow_T3
## Leaves 0.34340703
## Tot 0.58605625
## Root 0.60882466
## Stems 0.59138834
## LDA1_biomass_vs_treatment 0.97002406
## Area 0.78449689
## LDA1_area_vs_treatment 0.78449689
## Biostimulant_TimePointSumagrow_T3 1.00000000
## Biostimulant_TimePointControl_T3 0.51542821
## Biostimulant_TimePointInocucor_T3 0.51542821
## Biostimulant_TimePointPathway_T3 0.51542821
## Biostimulant_TimePointEndomaxx_T3 0.51542821
## SPAD.avg 0.71247714
## Biostimulant_TimePointSumagrow_T2 0.51542821
## Biostimulant_TimePointPathway_T2 0.51542821
## BiostimulantSumagrow 1.02689126
## Biostimulant_TimePointInocucor_T2 0.51542821
## Biostimulant_TimePointControl_T2 0.51542821
## BiostimulantPathway 0.45335192
## BiostimulantInocucor 0.45335192
## Biostimulant_TimePointEndomaxx_T2 0.51542821
## BiostimulantEndomaxx 0.45335192
## Biostimulant_TimePointInocucor_T1 0.51542821
## Biostimulant_TimePointSumagrow_T1 -0.05243893
## Biostimulant_TimePointPathway_T1 -0.05261428
## Biostimulant_TimePointControl_T1 -0.04379979
## Biostimulant_TimePointEndomaxx_T1 -0.01792098
## Biostimulant_TimePointPathway_T0 -0.02652675
## Biostimulant_TimePointSumagrow_T0 -0.03270557
## Biostimulant_TimePointInocucor_T0 -0.05243893
## Biostimulant_TimePointEndomaxx_T0 -0.02123148
## Biostimulant_TimePointControl_T3
## Leaves 0.39367886
## Tot 0.61139701
## Root 0.59845048
## Stems 0.61533062
## LDA1_biomass_vs_treatment 0.95106366
## Area 0.68032365
## LDA1_area_vs_treatment 0.68032365
## Biostimulant_TimePointSumagrow_T3 0.51542821
## Biostimulant_TimePointControl_T3 1.00000000
## Biostimulant_TimePointInocucor_T3 0.51542821
## Biostimulant_TimePointPathway_T3 0.51542821
## Biostimulant_TimePointEndomaxx_T3 0.51542821
## SPAD.avg 0.47076315
## Biostimulant_TimePointSumagrow_T2 0.51542821
## Biostimulant_TimePointPathway_T2 0.51542821
## BiostimulantSumagrow 0.45335192
## Biostimulant_TimePointInocucor_T2 0.51542821
## Biostimulant_TimePointControl_T2 0.51542821
## BiostimulantPathway 0.45335192
## BiostimulantInocucor 0.45335192
## Biostimulant_TimePointEndomaxx_T2 0.51542821
## BiostimulantEndomaxx 0.45335192
## Biostimulant_TimePointInocucor_T1 0.51542821
## Biostimulant_TimePointSumagrow_T1 0.71370780
## Biostimulant_TimePointPathway_T1 0.01843060
## Biostimulant_TimePointControl_T1 0.06427478
## Biostimulant_TimePointEndomaxx_T1 0.02945594
## Biostimulant_TimePointPathway_T0 0.07628471
## Biostimulant_TimePointSumagrow_T0 0.04773485
## Biostimulant_TimePointInocucor_T0 0.71370780
## Biostimulant_TimePointEndomaxx_T0 0.07993096
## Biostimulant_TimePointInocucor_T3
## Leaves 0.3535149
## Tot 0.5809237
## Root 0.5857597
## Stems 0.5837247
## LDA1_biomass_vs_treatment 0.9423330
## Area 0.7170768
## LDA1_area_vs_treatment 0.7170768
## Biostimulant_TimePointSumagrow_T3 0.5154282
## Biostimulant_TimePointControl_T3 0.5154282
## Biostimulant_TimePointInocucor_T3 1.0000000
## Biostimulant_TimePointPathway_T3 0.5154282
## Biostimulant_TimePointEndomaxx_T3 0.5154282
## SPAD.avg 0.5783072
## Biostimulant_TimePointSumagrow_T2 0.5154282
## Biostimulant_TimePointPathway_T2 0.5154282
## BiostimulantSumagrow 0.4533519
## Biostimulant_TimePointInocucor_T2 0.5154282
## Biostimulant_TimePointControl_T2 0.5154282
## BiostimulantPathway 0.4533519
## BiostimulantInocucor 1.0268913
## Biostimulant_TimePointEndomaxx_T2 0.5154282
## BiostimulantEndomaxx 0.4533519
## Biostimulant_TimePointInocucor_T1 0.5154282
## Biostimulant_TimePointSumagrow_T1 0.1119577
## Biostimulant_TimePointPathway_T1 0.5086011
## Biostimulant_TimePointControl_T1 0.7932143
## Biostimulant_TimePointEndomaxx_T1 0.7069257
## Biostimulant_TimePointPathway_T0 0.5488853
## Biostimulant_TimePointSumagrow_T0 0.6969154
## Biostimulant_TimePointInocucor_T0 0.1119577
## Biostimulant_TimePointEndomaxx_T0 0.5190482
## Biostimulant_TimePointPathway_T3
## Leaves 0.34334009
## Tot 0.58502053
## Root 0.58565901
## Stems 0.60121895
## LDA1_biomass_vs_treatment 0.94390155
## Area 0.71058866
## LDA1_area_vs_treatment 0.71058866
## Biostimulant_TimePointSumagrow_T3 0.51542821
## Biostimulant_TimePointControl_T3 0.51542821
## Biostimulant_TimePointInocucor_T3 0.51542821
## Biostimulant_TimePointPathway_T3 1.00000000
## Biostimulant_TimePointEndomaxx_T3 0.51542821
## SPAD.avg 0.53457190
## Biostimulant_TimePointSumagrow_T2 0.51542821
## Biostimulant_TimePointPathway_T2 0.51542821
## BiostimulantSumagrow 0.45335192
## Biostimulant_TimePointInocucor_T2 0.51542821
## Biostimulant_TimePointControl_T2 0.51542821
## BiostimulantPathway 1.02689126
## BiostimulantInocucor 0.45335192
## Biostimulant_TimePointEndomaxx_T2 0.51542821
## BiostimulantEndomaxx 0.45335192
## Biostimulant_TimePointInocucor_T1 0.51542821
## Biostimulant_TimePointSumagrow_T1 -0.02023767
## Biostimulant_TimePointPathway_T1 0.70258484
## Biostimulant_TimePointControl_T1 -0.05190465
## Biostimulant_TimePointEndomaxx_T1 -0.05205926
## Biostimulant_TimePointPathway_T0 -0.05260295
## Biostimulant_TimePointSumagrow_T0 -0.05233390
## Biostimulant_TimePointInocucor_T0 -0.02023767
## Biostimulant_TimePointEndomaxx_T0 -0.05261217
## Biostimulant_TimePointEndomaxx_T3
## Leaves 0.51651659
## Tot 0.32462232
## Root 0.58424893
## Stems 0.52209842
## LDA1_biomass_vs_treatment 0.94730194
## Area 0.76352779
## LDA1_area_vs_treatment 0.76352779
## Biostimulant_TimePointSumagrow_T3 0.51542821
## Biostimulant_TimePointControl_T3 0.51542821
## Biostimulant_TimePointInocucor_T3 0.51542821
## Biostimulant_TimePointPathway_T3 0.51542821
## Biostimulant_TimePointEndomaxx_T3 1.00000000
## SPAD.avg 0.53930425
## Biostimulant_TimePointSumagrow_T2 0.51542821
## Biostimulant_TimePointPathway_T2 0.51542821
## BiostimulantSumagrow 0.45335192
## Biostimulant_TimePointInocucor_T2 0.51542821
## Biostimulant_TimePointControl_T2 0.51542821
## BiostimulantPathway 0.45335192
## BiostimulantInocucor 0.45335192
## Biostimulant_TimePointEndomaxx_T2 0.51542821
## BiostimulantEndomaxx 1.02689126
## Biostimulant_TimePointInocucor_T1 0.51542821
## Biostimulant_TimePointSumagrow_T1 -0.05229159
## Biostimulant_TimePointPathway_T1 -0.05220574
## Biostimulant_TimePointControl_T1 -0.04365947
## Biostimulant_TimePointEndomaxx_T1 -0.01724260
## Biostimulant_TimePointPathway_T0 -0.02586869
## Biostimulant_TimePointSumagrow_T0 -0.03230415
## Biostimulant_TimePointInocucor_T0 -0.05229159
## Biostimulant_TimePointEndomaxx_T0 -0.02041589
## SPAD.avg
## Leaves 0.434335261
## Tot 0.155594941
## Root 0.352663355
## Stems 0.320624190
## LDA1_biomass_vs_treatment 0.695254139
## Area 1.111879922
## LDA1_area_vs_treatment 1.111879922
## Biostimulant_TimePointSumagrow_T3 0.712477142
## Biostimulant_TimePointControl_T3 0.470763146
## Biostimulant_TimePointInocucor_T3 0.578307153
## Biostimulant_TimePointPathway_T3 0.534571897
## Biostimulant_TimePointEndomaxx_T3 0.539304249
## SPAD.avg 1.000000000
## Biostimulant_TimePointSumagrow_T2 0.776597916
## Biostimulant_TimePointPathway_T2 0.691311566
## BiostimulantSumagrow 0.708351808
## Biostimulant_TimePointInocucor_T2 0.597704597
## Biostimulant_TimePointControl_T2 0.710552999
## BiostimulantPathway 0.639186188
## BiostimulantInocucor 0.518819309
## Biostimulant_TimePointEndomaxx_T2 0.514498402
## BiostimulantEndomaxx 0.405536168
## Biostimulant_TimePointInocucor_T1 0.754444267
## Biostimulant_TimePointSumagrow_T1 0.638669744
## Biostimulant_TimePointPathway_T1 0.233784243
## Biostimulant_TimePointControl_T1 0.389563933
## Biostimulant_TimePointEndomaxx_T1 0.280435362
## Biostimulant_TimePointPathway_T0 0.008915816
## Biostimulant_TimePointSumagrow_T0 -0.125088022
## Biostimulant_TimePointInocucor_T0 0.147441171
## Biostimulant_TimePointEndomaxx_T0 -0.223017694
## Biostimulant_TimePointSumagrow_T2
## Leaves 0.245530658
## Tot -0.003210153
## Root 0.198218581
## Stems 0.185812419
## LDA1_biomass_vs_treatment 0.547022353
## Area 0.729317284
## LDA1_area_vs_treatment 0.729317284
## Biostimulant_TimePointSumagrow_T3 0.515428211
## Biostimulant_TimePointControl_T3 0.515428211
## Biostimulant_TimePointInocucor_T3 0.515428211
## Biostimulant_TimePointPathway_T3 0.515428211
## Biostimulant_TimePointEndomaxx_T3 0.515428211
## SPAD.avg 0.776597916
## Biostimulant_TimePointSumagrow_T2 1.000000000
## Biostimulant_TimePointPathway_T2 0.515428211
## BiostimulantSumagrow 1.026891258
## Biostimulant_TimePointInocucor_T2 0.515428211
## Biostimulant_TimePointControl_T2 0.515428211
## BiostimulantPathway 0.453351923
## BiostimulantInocucor 0.453351923
## Biostimulant_TimePointEndomaxx_T2 0.515428211
## BiostimulantEndomaxx 0.453351923
## Biostimulant_TimePointInocucor_T1 0.515428211
## Biostimulant_TimePointSumagrow_T1 0.035169636
## Biostimulant_TimePointPathway_T1 0.695264952
## Biostimulant_TimePointControl_T1 0.610440398
## Biostimulant_TimePointEndomaxx_T1 0.762832520
## Biostimulant_TimePointPathway_T0 0.403323976
## Biostimulant_TimePointSumagrow_T0 0.850386420
## Biostimulant_TimePointInocucor_T0 0.035169636
## Biostimulant_TimePointEndomaxx_T0 0.309322388
## Biostimulant_TimePointPathway_T2
## Leaves 0.240931890
## Tot 0.006330031
## Root 0.215008640
## Stems 0.202540305
## LDA1_biomass_vs_treatment 0.566882017
## Area 0.712620257
## LDA1_area_vs_treatment 0.712620257
## Biostimulant_TimePointSumagrow_T3 0.515428211
## Biostimulant_TimePointControl_T3 0.515428211
## Biostimulant_TimePointInocucor_T3 0.515428211
## Biostimulant_TimePointPathway_T3 0.515428211
## Biostimulant_TimePointEndomaxx_T3 0.515428211
## SPAD.avg 0.691311566
## Biostimulant_TimePointSumagrow_T2 0.515428211
## Biostimulant_TimePointPathway_T2 1.000000000
## BiostimulantSumagrow 0.453351923
## Biostimulant_TimePointInocucor_T2 0.515428211
## Biostimulant_TimePointControl_T2 0.515428211
## BiostimulantPathway 1.026891258
## BiostimulantInocucor 0.453351923
## Biostimulant_TimePointEndomaxx_T2 0.515428211
## BiostimulantEndomaxx 0.453351923
## Biostimulant_TimePointInocucor_T1 0.515428211
## Biostimulant_TimePointSumagrow_T1 0.051659546
## Biostimulant_TimePointPathway_T1 0.858919594
## Biostimulant_TimePointControl_T1 -0.052556849
## Biostimulant_TimePointEndomaxx_T1 -0.052602140
## Biostimulant_TimePointPathway_T0 -0.052605175
## Biostimulant_TimePointSumagrow_T0 -0.052597425
## Biostimulant_TimePointInocucor_T0 0.051659546
## Biostimulant_TimePointEndomaxx_T0 -0.052606172
## BiostimulantSumagrow
## Leaves 0.22194174
## Tot 0.22515678
## Root 0.02630018
## Stems 0.22036781
## LDA1_biomass_vs_treatment 0.59647682
## Area 0.60320515
## LDA1_area_vs_treatment 0.60320515
## Biostimulant_TimePointSumagrow_T3 1.02689126
## Biostimulant_TimePointControl_T3 0.45335192
## Biostimulant_TimePointInocucor_T3 0.45335192
## Biostimulant_TimePointPathway_T3 0.45335192
## Biostimulant_TimePointEndomaxx_T3 0.45335192
## SPAD.avg 0.70835181
## Biostimulant_TimePointSumagrow_T2 1.02689126
## Biostimulant_TimePointPathway_T2 0.45335192
## BiostimulantSumagrow 1.00000000
## Biostimulant_TimePointInocucor_T2 0.45335192
## Biostimulant_TimePointControl_T2 0.45335192
## BiostimulantPathway 0.31805979
## BiostimulantInocucor 0.31805979
## Biostimulant_TimePointEndomaxx_T2 0.45335192
## BiostimulantEndomaxx 0.31805979
## Biostimulant_TimePointInocucor_T1 0.45335192
## Biostimulant_TimePointSumagrow_T1 0.45897637
## Biostimulant_TimePointPathway_T1 -0.01609344
## Biostimulant_TimePointControl_T1 -0.10566186
## Biostimulant_TimePointEndomaxx_T1 -0.07906615
## Biostimulant_TimePointPathway_T0 -0.08771212
## Biostimulant_TimePointSumagrow_T0 0.47932140
## Biostimulant_TimePointInocucor_T0 -0.11456297
## Biostimulant_TimePointEndomaxx_T0 -0.08221434
## Biostimulant_TimePointInocucor_T2
## Leaves 0.25237861
## Tot 0.20096196
## Root -0.06869508
## Stems 0.19067487
## LDA1_biomass_vs_treatment 0.48410639
## Area 0.72457336
## LDA1_area_vs_treatment 0.72457336
## Biostimulant_TimePointSumagrow_T3 0.51542821
## Biostimulant_TimePointControl_T3 0.51542821
## Biostimulant_TimePointInocucor_T3 0.51542821
## Biostimulant_TimePointPathway_T3 0.51542821
## Biostimulant_TimePointEndomaxx_T3 0.51542821
## SPAD.avg 0.59770460
## Biostimulant_TimePointSumagrow_T2 0.51542821
## Biostimulant_TimePointPathway_T2 0.51542821
## BiostimulantSumagrow 0.45335192
## Biostimulant_TimePointInocucor_T2 1.00000000
## Biostimulant_TimePointControl_T2 0.51542821
## BiostimulantPathway 0.45335192
## BiostimulantInocucor 1.02689126
## Biostimulant_TimePointEndomaxx_T2 0.51542821
## BiostimulantEndomaxx 0.45335192
## Biostimulant_TimePointInocucor_T1 0.51542821
## Biostimulant_TimePointSumagrow_T1 0.84326922
## Biostimulant_TimePointPathway_T1 -0.01723015
## Biostimulant_TimePointControl_T1 0.08445510
## Biostimulant_TimePointEndomaxx_T1 0.05172942
## Biostimulant_TimePointPathway_T0 0.08648990
## Biostimulant_TimePointSumagrow_T0 0.06700611
## Biostimulant_TimePointInocucor_T0 0.84326922
## Biostimulant_TimePointEndomaxx_T0 0.08880906
## Biostimulant_TimePointControl_T2
## Leaves 0.22641263
## Tot 0.17853275
## Root -0.08003617
## Stems 0.16562268
## LDA1_biomass_vs_treatment 0.47486454
## Area 0.73897053
## LDA1_area_vs_treatment 0.73897053
## Biostimulant_TimePointSumagrow_T3 0.51542821
## Biostimulant_TimePointControl_T3 0.51542821
## Biostimulant_TimePointInocucor_T3 0.51542821
## Biostimulant_TimePointPathway_T3 0.51542821
## Biostimulant_TimePointEndomaxx_T3 0.51542821
## SPAD.avg 0.71055300
## Biostimulant_TimePointSumagrow_T2 0.51542821
## Biostimulant_TimePointPathway_T2 0.51542821
## BiostimulantSumagrow 0.45335192
## Biostimulant_TimePointInocucor_T2 0.51542821
## Biostimulant_TimePointControl_T2 1.00000000
## BiostimulantPathway 0.45335192
## BiostimulantInocucor 0.45335192
## Biostimulant_TimePointEndomaxx_T2 0.51542821
## BiostimulantEndomaxx 0.45335192
## Biostimulant_TimePointInocucor_T1 0.51542821
## Biostimulant_TimePointSumagrow_T1 0.06255752
## Biostimulant_TimePointPathway_T1 0.12724239
## Biostimulant_TimePointControl_T1 0.70280105
## Biostimulant_TimePointEndomaxx_T1 0.86395099
## Biostimulant_TimePointPathway_T0 0.92310273
## Biostimulant_TimePointSumagrow_T0 0.89266416
## Biostimulant_TimePointInocucor_T0 0.06255752
## Biostimulant_TimePointEndomaxx_T0 0.93718225
## BiostimulantPathway BiostimulantInocucor
## Leaves 0.211222326 0.231592502
## Tot 0.225802627 0.218743512
## Root 0.020993415 0.197864978
## Stems 0.233099163 0.007143058
## LDA1_biomass_vs_treatment 0.592183861 0.549022196
## Area 0.545793957 0.596014664
## LDA1_area_vs_treatment 0.545793957 0.596014664
## Biostimulant_TimePointSumagrow_T3 0.453351923 0.453351923
## Biostimulant_TimePointControl_T3 0.453351923 0.453351923
## Biostimulant_TimePointInocucor_T3 0.453351923 1.026891258
## Biostimulant_TimePointPathway_T3 1.026891258 0.453351923
## Biostimulant_TimePointEndomaxx_T3 0.453351923 0.453351923
## SPAD.avg 0.639186188 0.518819309
## Biostimulant_TimePointSumagrow_T2 0.453351923 0.453351923
## Biostimulant_TimePointPathway_T2 1.026891258 0.453351923
## BiostimulantSumagrow 0.318059790 0.318059790
## Biostimulant_TimePointInocucor_T2 0.453351923 1.026891258
## Biostimulant_TimePointControl_T2 0.453351923 0.453351923
## BiostimulantPathway 1.000000000 0.318059790
## BiostimulantInocucor 0.318059790 1.000000000
## Biostimulant_TimePointEndomaxx_T2 0.453351923 0.453351923
## BiostimulantEndomaxx 0.318059790 0.318059790
## Biostimulant_TimePointInocucor_T1 0.453351923 1.026891258
## Biostimulant_TimePointSumagrow_T1 0.005733727 -0.114524032
## Biostimulant_TimePointPathway_T1 1.175358918 -0.111853048
## Biostimulant_TimePointControl_T1 -0.114584388 -0.105534803
## Biostimulant_TimePointEndomaxx_T1 -0.114696589 -0.078477527
## Biostimulant_TimePointPathway_T0 0.458858014 -0.087571573
## Biostimulant_TimePointSumagrow_T0 -0.114680397 -0.093943654
## Biostimulant_TimePointInocucor_T0 0.005733727 0.459015303
## Biostimulant_TimePointEndomaxx_T0 -0.114684520 -0.082079474
## Biostimulant_TimePointEndomaxx_T2
## Leaves 0.194268115
## Tot 0.182758507
## Root 0.164054286
## Stems -0.028229295
## LDA1_biomass_vs_treatment 0.515919081
## Area 0.695747215
## LDA1_area_vs_treatment 0.695747215
## Biostimulant_TimePointSumagrow_T3 0.515428211
## Biostimulant_TimePointControl_T3 0.515428211
## Biostimulant_TimePointInocucor_T3 0.515428211
## Biostimulant_TimePointPathway_T3 0.515428211
## Biostimulant_TimePointEndomaxx_T3 0.515428211
## SPAD.avg 0.514498402
## Biostimulant_TimePointSumagrow_T2 0.515428211
## Biostimulant_TimePointPathway_T2 0.515428211
## BiostimulantSumagrow 0.453351923
## Biostimulant_TimePointInocucor_T2 0.515428211
## Biostimulant_TimePointControl_T2 0.515428211
## BiostimulantPathway 0.453351923
## BiostimulantInocucor 0.453351923
## Biostimulant_TimePointEndomaxx_T2 1.000000000
## BiostimulantEndomaxx 1.026891258
## Biostimulant_TimePointInocucor_T1 0.515428211
## Biostimulant_TimePointSumagrow_T1 0.760285183
## Biostimulant_TimePointPathway_T1 0.003878776
## Biostimulant_TimePointControl_T1 0.132008612
## Biostimulant_TimePointEndomaxx_T1 0.057975685
## Biostimulant_TimePointPathway_T0 0.096080454
## Biostimulant_TimePointSumagrow_T0 0.086057018
## Biostimulant_TimePointInocucor_T0 0.760285183
## Biostimulant_TimePointEndomaxx_T0 0.093225514
## BiostimulantEndomaxx
## Leaves 0.16011999
## Tot 0.17525013
## Root 0.20207966
## Stems -0.03851243
## LDA1_biomass_vs_treatment 0.56393470
## Area 0.54720190
## LDA1_area_vs_treatment 0.54720190
## Biostimulant_TimePointSumagrow_T3 0.45335192
## Biostimulant_TimePointControl_T3 0.45335192
## Biostimulant_TimePointInocucor_T3 0.45335192
## Biostimulant_TimePointPathway_T3 0.45335192
## Biostimulant_TimePointEndomaxx_T3 1.02689126
## SPAD.avg 0.40553617
## Biostimulant_TimePointSumagrow_T2 0.45335192
## Biostimulant_TimePointPathway_T2 0.45335192
## BiostimulantSumagrow 0.31805979
## Biostimulant_TimePointInocucor_T2 0.45335192
## Biostimulant_TimePointControl_T2 0.45335192
## BiostimulantPathway 0.31805979
## BiostimulantInocucor 0.31805979
## Biostimulant_TimePointEndomaxx_T2 1.02689126
## BiostimulantEndomaxx 1.00000000
## Biostimulant_TimePointInocucor_T1 0.45335192
## Biostimulant_TimePointSumagrow_T1 -0.03622074
## Biostimulant_TimePointPathway_T1 -0.09243608
## Biostimulant_TimePointControl_T1 0.60311508
## Biostimulant_TimePointEndomaxx_T1 1.23362256
## Biostimulant_TimePointPathway_T0 0.76480168
## Biostimulant_TimePointSumagrow_T0 0.85753219
## Biostimulant_TimePointInocucor_T0 -0.03622074
## Biostimulant_TimePointEndomaxx_T0 1.27842588
## Biostimulant_TimePointInocucor_T1
## Leaves 0.09170286
## Tot 0.07834586
## Root 0.08324646
## Stems -0.14461099
## LDA1_biomass_vs_treatment 0.43846561
## Area 0.63552361
## LDA1_area_vs_treatment 0.63552361
## Biostimulant_TimePointSumagrow_T3 0.51542821
## Biostimulant_TimePointControl_T3 0.51542821
## Biostimulant_TimePointInocucor_T3 0.51542821
## Biostimulant_TimePointPathway_T3 0.51542821
## Biostimulant_TimePointEndomaxx_T3 0.51542821
## SPAD.avg 0.75444427
## Biostimulant_TimePointSumagrow_T2 0.51542821
## Biostimulant_TimePointPathway_T2 0.51542821
## BiostimulantSumagrow 0.45335192
## Biostimulant_TimePointInocucor_T2 0.51542821
## Biostimulant_TimePointControl_T2 0.51542821
## BiostimulantPathway 0.45335192
## BiostimulantInocucor 1.02689126
## Biostimulant_TimePointEndomaxx_T2 0.51542821
## BiostimulantEndomaxx 0.45335192
## Biostimulant_TimePointInocucor_T1 1.00000000
## Biostimulant_TimePointSumagrow_T1 -0.03505166
## Biostimulant_TimePointPathway_T1 0.06292161
## Biostimulant_TimePointControl_T1 -0.05250850
## Biostimulant_TimePointEndomaxx_T1 -0.05261211
## Biostimulant_TimePointPathway_T0 -0.05262428
## Biostimulant_TimePointSumagrow_T0 -0.05260559
## Biostimulant_TimePointInocucor_T0 -0.03505166
## Biostimulant_TimePointEndomaxx_T0 -0.05262623
## Biostimulant_TimePointSumagrow_T1
## Leaves 0.699182697
## Tot 0.625862955
## Root 0.676622855
## Stems 0.556747101
## LDA1_biomass_vs_treatment 0.771902116
## Area 0.083816638
## LDA1_area_vs_treatment 0.243999664
## Biostimulant_TimePointSumagrow_T3 -0.052438928
## Biostimulant_TimePointControl_T3 0.713707800
## Biostimulant_TimePointInocucor_T3 0.111957723
## Biostimulant_TimePointPathway_T3 -0.020237671
## Biostimulant_TimePointEndomaxx_T3 -0.052291586
## SPAD.avg 0.638669744
## Biostimulant_TimePointSumagrow_T2 0.035169636
## Biostimulant_TimePointPathway_T2 0.051659546
## BiostimulantSumagrow 0.458976367
## Biostimulant_TimePointInocucor_T2 0.843269216
## Biostimulant_TimePointControl_T2 0.062557522
## BiostimulantPathway 0.005733727
## BiostimulantInocucor -0.114524032
## Biostimulant_TimePointEndomaxx_T2 0.760285183
## BiostimulantEndomaxx -0.036220738
## Biostimulant_TimePointInocucor_T1 -0.035051664
## Biostimulant_TimePointSumagrow_T1 1.000000000
## Biostimulant_TimePointPathway_T1 -0.052631579
## Biostimulant_TimePointControl_T1 -0.052631579
## Biostimulant_TimePointEndomaxx_T1 -0.052631579
## Biostimulant_TimePointPathway_T0 -0.052631579
## Biostimulant_TimePointSumagrow_T0 -0.052631579
## Biostimulant_TimePointInocucor_T0 -0.052631579
## Biostimulant_TimePointEndomaxx_T0 -0.052631579
## Biostimulant_TimePointPathway_T1
## Leaves -0.060371369
## Tot 0.450450390
## Root 0.304314809
## Stems -0.022550517
## LDA1_biomass_vs_treatment -0.106270215
## Area 0.132605457
## LDA1_area_vs_treatment 0.302496746
## Biostimulant_TimePointSumagrow_T3 -0.052614282
## Biostimulant_TimePointControl_T3 0.018430601
## Biostimulant_TimePointInocucor_T3 0.508601066
## Biostimulant_TimePointPathway_T3 0.702584843
## Biostimulant_TimePointEndomaxx_T3 -0.052205736
## SPAD.avg 0.233784243
## Biostimulant_TimePointSumagrow_T2 0.695264952
## Biostimulant_TimePointPathway_T2 0.858919594
## BiostimulantSumagrow -0.016093440
## Biostimulant_TimePointInocucor_T2 -0.017230150
## Biostimulant_TimePointControl_T2 0.127242394
## BiostimulantPathway 1.175358918
## BiostimulantInocucor -0.111853048
## Biostimulant_TimePointEndomaxx_T2 0.003878776
## BiostimulantEndomaxx -0.092436078
## Biostimulant_TimePointInocucor_T1 0.062921609
## Biostimulant_TimePointSumagrow_T1 -0.052631579
## Biostimulant_TimePointPathway_T1 1.000000000
## Biostimulant_TimePointControl_T1 -0.037879867
## Biostimulant_TimePointEndomaxx_T1 0.185867693
## Biostimulant_TimePointPathway_T0 0.073345063
## Biostimulant_TimePointSumagrow_T0 0.037071373
## Biostimulant_TimePointInocucor_T0 -0.052631579
## Biostimulant_TimePointEndomaxx_T0 0.113607469
## Biostimulant_TimePointControl_T1
## Leaves 0.44454594
## Tot 0.68610099
## Root 0.84565505
## Stems 0.75548059
## LDA1_biomass_vs_treatment 0.01715083
## Area 0.88853235
## LDA1_area_vs_treatment 0.01096427
## Biostimulant_TimePointSumagrow_T3 -0.04379979
## Biostimulant_TimePointControl_T3 0.06427478
## Biostimulant_TimePointInocucor_T3 0.79321433
## Biostimulant_TimePointPathway_T3 -0.05190465
## Biostimulant_TimePointEndomaxx_T3 -0.04365947
## SPAD.avg 0.38956393
## Biostimulant_TimePointSumagrow_T2 0.61044040
## Biostimulant_TimePointPathway_T2 -0.05255685
## BiostimulantSumagrow -0.10566186
## Biostimulant_TimePointInocucor_T2 0.08445510
## Biostimulant_TimePointControl_T2 0.70280105
## BiostimulantPathway -0.11458439
## BiostimulantInocucor -0.10553480
## Biostimulant_TimePointEndomaxx_T2 0.13200861
## BiostimulantEndomaxx 0.60311508
## Biostimulant_TimePointInocucor_T1 -0.05250850
## Biostimulant_TimePointSumagrow_T1 -0.05263158
## Biostimulant_TimePointPathway_T1 -0.03787987
## Biostimulant_TimePointControl_T1 1.00000000
## Biostimulant_TimePointEndomaxx_T1 -0.05263158
## Biostimulant_TimePointPathway_T0 -0.05263158
## Biostimulant_TimePointSumagrow_T0 -0.05263158
## Biostimulant_TimePointInocucor_T0 -0.05263158
## Biostimulant_TimePointEndomaxx_T0 -0.05263158
## Biostimulant_TimePointEndomaxx_T1
## Leaves 0.53233412
## Tot 0.78757103
## Root 0.68139120
## Stems 0.77009775
## LDA1_biomass_vs_treatment -0.05805714
## Area 0.58601563
## LDA1_area_vs_treatment -0.02740427
## Biostimulant_TimePointSumagrow_T3 -0.01792098
## Biostimulant_TimePointControl_T3 0.02945594
## Biostimulant_TimePointInocucor_T3 0.70692574
## Biostimulant_TimePointPathway_T3 -0.05205926
## Biostimulant_TimePointEndomaxx_T3 -0.01724260
## SPAD.avg 0.28043536
## Biostimulant_TimePointSumagrow_T2 0.76283252
## Biostimulant_TimePointPathway_T2 -0.05260214
## BiostimulantSumagrow -0.07906615
## Biostimulant_TimePointInocucor_T2 0.05172942
## Biostimulant_TimePointControl_T2 0.86395099
## BiostimulantPathway -0.11469659
## BiostimulantInocucor -0.07847753
## Biostimulant_TimePointEndomaxx_T2 0.05797568
## BiostimulantEndomaxx 1.23362256
## Biostimulant_TimePointInocucor_T1 -0.05261211
## Biostimulant_TimePointSumagrow_T1 -0.05263158
## Biostimulant_TimePointPathway_T1 0.18586769
## Biostimulant_TimePointControl_T1 -0.05263158
## Biostimulant_TimePointEndomaxx_T1 1.00000000
## Biostimulant_TimePointPathway_T0 -0.05263158
## Biostimulant_TimePointSumagrow_T0 -0.05263158
## Biostimulant_TimePointInocucor_T0 -0.05263158
## Biostimulant_TimePointEndomaxx_T0 -0.05263158
## Biostimulant_TimePointPathway_T0
## Leaves 0.675264423
## Tot 0.665134628
## Root 0.618048832
## Stems 0.583524966
## LDA1_biomass_vs_treatment -0.083451608
## Area 0.044837995
## LDA1_area_vs_treatment -0.340011255
## Biostimulant_TimePointSumagrow_T3 -0.026526753
## Biostimulant_TimePointControl_T3 0.076284710
## Biostimulant_TimePointInocucor_T3 0.548885265
## Biostimulant_TimePointPathway_T3 -0.052602952
## Biostimulant_TimePointEndomaxx_T3 -0.025868689
## SPAD.avg 0.008915816
## Biostimulant_TimePointSumagrow_T2 0.403323976
## Biostimulant_TimePointPathway_T2 -0.052605175
## BiostimulantSumagrow -0.087712117
## Biostimulant_TimePointInocucor_T2 0.086489902
## Biostimulant_TimePointControl_T2 0.923102735
## BiostimulantPathway 0.458858014
## BiostimulantInocucor -0.087571573
## Biostimulant_TimePointEndomaxx_T2 0.096080454
## BiostimulantEndomaxx 0.764801678
## Biostimulant_TimePointInocucor_T1 -0.052624276
## Biostimulant_TimePointSumagrow_T1 -0.052631579
## Biostimulant_TimePointPathway_T1 0.073345063
## Biostimulant_TimePointControl_T1 -0.052631579
## Biostimulant_TimePointEndomaxx_T1 -0.052631579
## Biostimulant_TimePointPathway_T0 1.000000000
## Biostimulant_TimePointSumagrow_T0 -0.052631579
## Biostimulant_TimePointInocucor_T0 -0.052631579
## Biostimulant_TimePointEndomaxx_T0 -0.052631579
## Biostimulant_TimePointSumagrow_T0
## Leaves 0.45171076
## Tot 0.73256934
## Root 0.68105724
## Stems 0.69690528
## LDA1_biomass_vs_treatment -0.08708204
## Area 0.38003179
## LDA1_area_vs_treatment -0.33501550
## Biostimulant_TimePointSumagrow_T3 -0.03270557
## Biostimulant_TimePointControl_T3 0.04773485
## Biostimulant_TimePointInocucor_T3 0.69691538
## Biostimulant_TimePointPathway_T3 -0.05233390
## Biostimulant_TimePointEndomaxx_T3 -0.03230415
## SPAD.avg -0.12508802
## Biostimulant_TimePointSumagrow_T2 0.85038642
## Biostimulant_TimePointPathway_T2 -0.05259743
## BiostimulantSumagrow 0.47932140
## Biostimulant_TimePointInocucor_T2 0.06700611
## Biostimulant_TimePointControl_T2 0.89266416
## BiostimulantPathway -0.11468040
## BiostimulantInocucor -0.09394365
## Biostimulant_TimePointEndomaxx_T2 0.08605702
## BiostimulantEndomaxx 0.85753219
## Biostimulant_TimePointInocucor_T1 -0.05260559
## Biostimulant_TimePointSumagrow_T1 -0.05263158
## Biostimulant_TimePointPathway_T1 0.03707137
## Biostimulant_TimePointControl_T1 -0.05263158
## Biostimulant_TimePointEndomaxx_T1 -0.05263158
## Biostimulant_TimePointPathway_T0 -0.05263158
## Biostimulant_TimePointSumagrow_T0 1.00000000
## Biostimulant_TimePointInocucor_T0 -0.05263158
## Biostimulant_TimePointEndomaxx_T0 -0.05263158
## Biostimulant_TimePointInocucor_T0
## Leaves 0.583503928
## Tot 0.550206651
## Root 0.607035263
## Stems 0.510865654
## LDA1_biomass_vs_treatment 0.709799496
## Area -0.259701868
## LDA1_area_vs_treatment -0.099518842
## Biostimulant_TimePointSumagrow_T3 -0.052438928
## Biostimulant_TimePointControl_T3 0.713707800
## Biostimulant_TimePointInocucor_T3 0.111957723
## Biostimulant_TimePointPathway_T3 -0.020237671
## Biostimulant_TimePointEndomaxx_T3 -0.052291586
## SPAD.avg 0.147441171
## Biostimulant_TimePointSumagrow_T2 0.035169636
## Biostimulant_TimePointPathway_T2 0.051659546
## BiostimulantSumagrow -0.114562967
## Biostimulant_TimePointInocucor_T2 0.843269216
## Biostimulant_TimePointControl_T2 0.062557522
## BiostimulantPathway 0.005733727
## BiostimulantInocucor 0.459015303
## Biostimulant_TimePointEndomaxx_T2 0.760285183
## BiostimulantEndomaxx -0.036220738
## Biostimulant_TimePointInocucor_T1 -0.035051664
## Biostimulant_TimePointSumagrow_T1 -0.052631579
## Biostimulant_TimePointPathway_T1 -0.052631579
## Biostimulant_TimePointControl_T1 -0.052631579
## Biostimulant_TimePointEndomaxx_T1 -0.052631579
## Biostimulant_TimePointPathway_T0 -0.052631579
## Biostimulant_TimePointSumagrow_T0 -0.052631579
## Biostimulant_TimePointInocucor_T0 1.000000000
## Biostimulant_TimePointEndomaxx_T0 -0.052631579
## Biostimulant_TimePointEndomaxx_T0
## Leaves 0.72625469
## Tot 0.62420587
## Root 0.59059149
## Stems 0.56503191
## LDA1_biomass_vs_treatment -0.08952663
## Area -0.02269113
## LDA1_area_vs_treatment -0.33401142
## Biostimulant_TimePointSumagrow_T3 -0.02123148
## Biostimulant_TimePointControl_T3 0.07993096
## Biostimulant_TimePointInocucor_T3 0.51904818
## Biostimulant_TimePointPathway_T3 -0.05261217
## Biostimulant_TimePointEndomaxx_T3 -0.02041589
## SPAD.avg -0.22301769
## Biostimulant_TimePointSumagrow_T2 0.30932239
## Biostimulant_TimePointPathway_T2 -0.05260617
## BiostimulantSumagrow -0.08221434
## Biostimulant_TimePointInocucor_T2 0.08880906
## Biostimulant_TimePointControl_T2 0.93718225
## BiostimulantPathway -0.11468452
## BiostimulantInocucor -0.08207947
## Biostimulant_TimePointEndomaxx_T2 0.09322551
## BiostimulantEndomaxx 1.27842588
## Biostimulant_TimePointInocucor_T1 -0.05262623
## Biostimulant_TimePointSumagrow_T1 -0.05263158
## Biostimulant_TimePointPathway_T1 0.11360747
## Biostimulant_TimePointControl_T1 -0.05263158
## Biostimulant_TimePointEndomaxx_T1 -0.05263158
## Biostimulant_TimePointPathway_T0 -0.05263158
## Biostimulant_TimePointSumagrow_T0 -0.05263158
## Biostimulant_TimePointInocucor_T0 -0.05263158
## Biostimulant_TimePointEndomaxx_T0 1.00000000Tukey’s HSD
library(agricolae)
formula <- lda_variables %>%
dplyr::select(-X.SampleID, -Biostimulant, -TimePoint, -Rep, -Biostimulant_TimePoint) %>%
colnames(.) %>%
paste(collapse = '+') %>%
paste0("~Biostimulant*TimePoint")
model <- aov(formula = as.formula(formula),data = lda_variables)
HSD.test(model, c('Biostimulant','TimePoint'), group=TRUE, console = TRUE)##
## Study: model ~ c("Biostimulant", "TimePoint")
##
## HSD Test for Area + SPAD.avg + Leaves + Stems + Root + Tot + LDA1_area_vs_treatment + LDA1_biomass_vs_treatment
##
## Mean Square Error: 59860423
##
## Biostimulant:TimePoint, means
##
## Area...SPAD.avg...Leaves...Stems...Root...Tot...LDA1_area_vs_treatment...LDA1_biomass_vs_treatment
## Control:T0 1423.980
## Control:T1 11221.281
## Control:T2 27919.769
## Control:T3 90523.541
## Endomaxx:T0 1233.455
## Endomaxx:T1 10655.031
## Endomaxx:T2 28458.386
## Endomaxx:T3 79857.140
## Inocucor:T0 1351.726
## Inocucor:T1 13246.916
## Inocucor:T2 31142.004
## Inocucor:T3 86171.488
## Pathway:T0 1366.712
## Pathway:T1 11859.052
## Pathway:T2 33673.238
## Pathway:T3 86753.210
## Sumagrow:T0 1435.490
## Sumagrow:T1 12869.290
## Sumagrow:T2 32320.106
## Sumagrow:T3 87026.510
## std r Min Max
## Control:T0 209.6647 6 1127.5490 1753.384
## Control:T1 1608.8545 6 9160.6688 13159.208
## Control:T2 11827.2449 6 6182.1711 38148.368
## Control:T3 11282.4934 6 71122.3143 103970.407
## Endomaxx:T0 336.6803 6 749.9229 1716.401
## Endomaxx:T1 2569.9170 6 6252.4763 13009.854
## Endomaxx:T2 8004.2091 6 12648.9982 35096.270
## Endomaxx:T3 6390.8685 6 68436.2083 87294.333
## Inocucor:T0 164.4797 6 1149.8716 1536.325
## Inocucor:T1 757.6288 6 12368.0014 14399.580
## Inocucor:T2 3135.2540 6 27487.8944 35814.160
## Inocucor:T3 22391.1987 6 64551.6091 126706.771
## Pathway:T0 100.4050 6 1244.5531 1510.122
## Pathway:T1 2537.1205 6 8119.7624 15480.086
## Pathway:T2 7134.4230 6 26832.5593 46420.858
## Pathway:T3 10122.5326 6 74033.4369 104154.662
## Sumagrow:T0 312.1367 6 892.4997 1730.269
## Sumagrow:T1 731.6217 6 11765.6893 13642.374
## Sumagrow:T2 8501.1937 6 22234.4478 42830.223
## Sumagrow:T3 8441.3829 6 75258.8619 98553.018
##
## Alpha: 0.05 ; DF Error: 100
## Critical Value of Studentized Range: 5.148913
##
## Minimun Significant Difference: 16263.34
##
## Treatments with the same letter are not significantly different.
##
## Area + SPAD.avg + Leaves + Stems + Root + Tot + LDA1_area_vs_treatment + LDA1_biomass_vs_treatment
## Control:T3 90523.541
## Sumagrow:T3 87026.510
## Pathway:T3 86753.210
## Inocucor:T3 86171.488
## Endomaxx:T3 79857.140
## Pathway:T2 33673.238
## Sumagrow:T2 32320.106
## Inocucor:T2 31142.004
## Endomaxx:T2 28458.386
## Control:T2 27919.769
## Inocucor:T1 13246.916
## Sumagrow:T1 12869.290
## Pathway:T1 11859.052
## Control:T1 11221.281
## Endomaxx:T1 10655.031
## Sumagrow:T0 1435.490
## Control:T0 1423.980
## Pathway:T0 1366.712
## Inocucor:T0 1351.726
## Endomaxx:T0 1233.455
## groups
## Control:T3 a
## Sumagrow:T3 a
## Pathway:T3 a
## Inocucor:T3 a
## Endomaxx:T3 a
## Pathway:T2 b
## Sumagrow:T2 b
## Inocucor:T2 b
## Endomaxx:T2 bc
## Control:T2 bcd
## Inocucor:T1 cde
## Sumagrow:T1 cde
## Pathway:T1 de
## Control:T1 e
## Endomaxx:T1 e
## Sumagrow:T0 e
## Control:T0 e
## Pathway:T0 e
## Inocucor:T0 e
## Endomaxx:T0 ereadr::write_tsv(lda_variables,'mappings_LDA.txt')Phyloseq
Import files
library(phyloseq)
library(vegan)## Loading required package: permute## Loading required package: lattice## This is vegan 2.5-2library(ape)##
## Attaching package: 'ape'## The following object is masked from 'package:agricolae':
##
## consensuslibrary(dummies)## dummies-1.5.6 provided by Decision Patterns#File Paths
biom_path <- file.path('merged/biom/table_wo_chl_mit.biom')
tree_path <- file.path('merged/biom/tree.nwk')
DESEq2_path <- file.path('merged/biom/DESeq2_w_tax.biom')
map_path <- file.path('mappings_LDA.txt')
#Import to phyloseq table and merge into phyloseq objects
table <- import_biom(BIOMfilename = biom_path,
#refseqfilename = repseqfile,
parseFunction = parse_taxonomy_default,
parallel = T)## Warning in strsplit(conditionMessage(e), "\n"): input string 1 is invalid
## in this localetax_table(table) <-tax_table(table)[,1:7]
DESEq2.table <- import_biom(BIOMfilename = DESEq2_path,
#refseqfilename = repseqfile,
parseFunction = parse_taxonomy_default,
parallel = T)## Warning in strsplit(conditionMessage(e), "\n"): input string 1 is invalid
## in this localetax_table(DESEq2.table) <-tax_table(DESEq2.table)[,1:7]
metadata <- import_qiime_sample_data(map_path)
tree <- read_tree(tree_path)
phylobj <- merge_phyloseq(table, metadata, tree)
DESEq2.phylobj <- merge_phyloseq(DESEq2.table, metadata, tree)
#Adjust taxonomy names (to harmonize betwen UNITE and SILVA databases)
tax_table(phylobj) <- gsub(".*__", "", tax_table(phylobj))
colnames(tax_table(phylobj)) <- c("Kingdom", "Phylum", "Class",
"Order", "Family", "Genus", "Species")
tax_table(DESEq2.phylobj) <- gsub(".*__", "", tax_table(DESEq2.phylobj))
colnames(tax_table(DESEq2.phylobj)) <- c("Kingdom", "Phylum", "Class",
"Order", "Family", "Genus", "Species")
#Log-transform sample counts (if needed)
pslog <- transform_sample_counts(phylobj,function(x){log(1 + x)})Phylum barplots
phyloglom <- tax_glom(phylobj, 'Phylum')
phylorel <- transform_sample_counts(phyloglom, function(x){100*x/sum(x)})
phylorel %>%
filter_taxa(function(x){ mean(x) > 1}, TRUE) %>%
psmelt()%>%
group_by(Biostimulant, TimePoint, Phylum) %>%
summarize(mean=mean(Abundance)) %>%
ggplot() +
aes(x=Biostimulant, y=mean, fill=Phylum, color=Phylum) +
geom_bar(stat='identity') +
facet_grid(.~TimePoint) +
ylab('Relative abundance') +
theme_igray() +
theme(axis.text.x = element_text(angle=90, vjust=0.5)) +
scale_fill_pander() + scale_color_pander()
Alpha Diveristy
plot_richness(phylobj,
x = "Biostimulant",
color = "TimePoint",
measures = c('Observed', 'Chao1', 'Shannon')) +
geom_boxplot(alpha=.9) + theme_igray() +
theme(axis.text.x = element_text(angle=90, vjust=0.5))+
scale_color_pander() + scale_fill_pander()## Warning: Removed 240 rows containing missing values (geom_errorbar).
##Rarefaction plots
source('https://raw.githubusercontent.com/mahendra-mariadassou/phyloseq-extended/master/R/richness.R')## Loading required package: parallelggrare(phylobj, step = 100, color = "Biostimulant", label = "Sample", se = FALSE) +
facet_wrap(~TimePoint) + guides(label=FALSE) + theme_igray() + scale_color_pander()## rarefying sample B2CT0-1
## rarefying sample B2CT0-2
## rarefying sample B2CT0-3
## rarefying sample B2CT0-4
## rarefying sample B2CT0-5
## rarefying sample B2CT0-6
## rarefying sample B2CT1-1
## rarefying sample B2CT1-2
## rarefying sample B2CT1-3
## rarefying sample B2CT1-4
## rarefying sample B2CT1-5
## rarefying sample B2CT1-6
## rarefying sample B2CT2-1
## rarefying sample B2CT2-2
## rarefying sample B2CT2-3
## rarefying sample B2CT2-4
## rarefying sample B2CT2-5
## rarefying sample B2CT2-6
## rarefying sample B2CT3-1
## rarefying sample B2CT3-2
## rarefying sample B2CT3-3
## rarefying sample B2CT3-4
## rarefying sample B2CT3-5
## rarefying sample B2CT3-6
## rarefying sample B2ET0-1
## rarefying sample B2ET0-2
## rarefying sample B2ET0-3
## rarefying sample B2ET0-4
## rarefying sample B2ET0-5
## rarefying sample B2ET0-6
## rarefying sample B2ET1-1
## rarefying sample B2ET1-2
## rarefying sample B2ET1-3
## rarefying sample B2ET1-4
## rarefying sample B2ET1-5
## rarefying sample B2ET1-6
## rarefying sample B2ET2-1
## rarefying sample B2ET2-2
## rarefying sample B2ET2-3
## rarefying sample B2ET2-4
## rarefying sample B2ET2-5
## rarefying sample B2ET2-6
## rarefying sample B2ET3-1
## rarefying sample B2ET3-2
## rarefying sample B2ET3-3
## rarefying sample B2ET3-4
## rarefying sample B2ET3-5
## rarefying sample B2ET3-6
## rarefying sample B2IT0-1
## rarefying sample B2IT0-2
## rarefying sample B2IT0-3
## rarefying sample B2IT0-4
## rarefying sample B2IT0-5
## rarefying sample B2IT0-6
## rarefying sample B2IT1-1
## rarefying sample B2IT1-2
## rarefying sample B2IT1-3
## rarefying sample B2IT1-4
## rarefying sample B2IT1-5
## rarefying sample B2IT1-6
## rarefying sample B2IT2-1
## rarefying sample B2IT2-2
## rarefying sample B2IT2-3
## rarefying sample B2IT2-4
## rarefying sample B2IT2-5
## rarefying sample B2IT2-6
## rarefying sample B2IT3-1
## rarefying sample B2IT3-2
## rarefying sample B2IT3-3
## rarefying sample B2IT3-4
## rarefying sample B2IT3-5
## rarefying sample B2IT3-6
## rarefying sample B2PT0-1
## rarefying sample B2PT0-2
## rarefying sample B2PT0-3
## rarefying sample B2PT0-4
## rarefying sample B2PT0-5
## rarefying sample B2PT0-6
## rarefying sample B2PT1-1
## rarefying sample B2PT1-2
## rarefying sample B2PT1-3
## rarefying sample B2PT1-4
## rarefying sample B2PT1-5
## rarefying sample B2PT1-6
## rarefying sample B2PT2-1
## rarefying sample B2PT2-2
## rarefying sample B2PT2-3
## rarefying sample B2PT2-4
## rarefying sample B2PT2-5
## rarefying sample B2PT2-6
## rarefying sample B2PT3-1
## rarefying sample B2PT3-2
## rarefying sample B2PT3-3
## rarefying sample B2PT3-4
## rarefying sample B2PT3-5
## rarefying sample B2PT3-6
## rarefying sample B2ST0-1
## rarefying sample B2ST0-2
## rarefying sample B2ST0-3
## rarefying sample B2ST0-4
## rarefying sample B2ST0-5
## rarefying sample B2ST0-6
## rarefying sample B2ST1-1
## rarefying sample B2ST1-2
## rarefying sample B2ST1-3
## rarefying sample B2ST1-4
## rarefying sample B2ST1-5
## rarefying sample B2ST1-6
## rarefying sample B2ST2-1
## rarefying sample B2ST2-2
## rarefying sample B2ST2-3
## rarefying sample B2ST2-4
## rarefying sample B2ST2-5
## rarefying sample B2ST2-6
## rarefying sample B2ST3-1
## rarefying sample B2ST3-2
## rarefying sample B2ST3-3
## rarefying sample B2ST3-4
## rarefying sample B2ST3-5
## rarefying sample B2ST3-6
PCoA with Bray distances
PCoA.bray.ord<- ordinate(DESEq2.phylobj, "PCoA", distance = "bray")
PCoA.bray.plot <- plot_ordination(physeq = pslog,
ordination = PCoA.bray.ord,
color = "Biostimulant",
shape = "TimePoint",
title = "PCoA (Bray-Curtis distances)")
PCoA.bray.plot +geom_point(size=3) + scale_color_pander()
##PCoA with Unifrac distances
PCoA.bray.ord<- ordinate(pslog, "PCoA", distance = "unifrac")
PCoA.bray.plot <- plot_ordination(physeq = pslog,
ordination = PCoA.bray.ord,
color = "Biostimulant",
shape = "TimePoint",
title = "PCoA (Unifrac distances)")
PCoA.bray.plot +geom_point(size=3) + scale_color_pander()
If you want to run PCoA with ALL possible distances (adapted from Phyloseq website https://joey711.github.io/phyloseq/distance.html) (requires a lot of memory)
dist_methods <- unlist(distanceMethodList)
dist_methods = dist_methods[-which(dist_methods=="ANY")]
plist <- vector("list", length(dist_methods))
names(plist) = dist_methods
for( i in dist_methods ){
# Calculate distance matrix
iDist <- distance(pslog, method=i)
# Calculate ordination
iMDS <- ordinate(pslog, "PCoA", distance=iDist)
## Make plot
# Don't carry over previous plot (if error, p will be blank)
p <- NULL
# Create plot, store as temp variable, p
p <- plot_ordination(pslog, iMDS, color="Biostimulant", shape="TimePoint")
# Add title to each plot
p <- p + ggtitle(paste("PCoA using distance method ", i, sep=""))
# Save the graphic to file.
plist[[i]] = p
}
df = ldply(plist, function(x) x$data)
names(df)[1] <- "distance"
p = ggplot(df, aes(Axis.1, Axis.2, color=Biostimulant, shape=TimePoint))
p = p + geom_point(size=3, alpha=0.5)
p = p + facet_wrap(~distance, scales="free")
p = p + ggtitle("PCoA on various distance metrics")
pNMDS
NMDS_bray_b1 <- ordinate(DESEq2.phylobj, method = "NMDS", distance = "bray")## Wisconsin double standardization
## Run 0 stress 0.03550533
## Run 1 stress 0.04837901
## Run 2 stress 0.03673057
## Run 3 stress 0.03601908
## Run 4 stress 0.03610428
## Run 5 stress 0.05136521
## Run 6 stress 0.03624341
## Run 7 stress 0.03911558
## Run 8 stress 0.04939574
## Run 9 stress 0.03635159
## Run 10 stress 0.04983131
## Run 11 stress 0.0365035
## Run 12 stress 0.04808414
## Run 13 stress 0.03634461
## Run 14 stress 0.03734913
## Run 15 stress 0.05002115
## Run 16 stress 0.05185519
## Run 17 stress 0.03556772
## ... Procrustes: rmse 0.001972417 max resid 0.01388614
## Run 18 stress 0.03580298
## ... Procrustes: rmse 0.0062454 max resid 0.04616961
## Run 19 stress 0.05155175
## Run 20 stress 0.05777514
## *** No convergence -- monoMDS stopping criteria:
## 17: no. of iterations >= maxit
## 3: stress ratio > sratmaxstressplot(NMDS_bray_b1)
plot_NMDS_bray_b1 <- plot_ordination(pslog,
NMDS_bray_b1,
type="samples",
color="Biostimulant",
shape="TimePoint")
plot_NMDS_bray_b1 + ggtitle("NMDS using Bray-Curtis distances") + scale_color_pander()
##PERMANOVA
DESEq2.df = as(sample_data(DESEq2.phylobj), "data.frame")
DESEq2.distbray = distance(DESEq2.phylobj, "bray")
adonis(DESEq2.distbray ~ TimePoint*Biostimulant, DESEq2.df)##
## Call:
## adonis(formula = DESEq2.distbray ~ TimePoint * Biostimulant, data = DESEq2.df)
##
## Permutation: free
## Number of permutations: 999
##
## Terms added sequentially (first to last)
##
## Df SumsOfSqs MeanSqs F.Model R2 Pr(>F)
## TimePoint 3 0.024838 0.0082795 5.7980 0.13002 0.001 ***
## Biostimulant 4 0.005316 0.0013289 0.9306 0.02783 0.653
## TimePoint:Biostimulant 12 0.018086 0.0015072 1.0555 0.09467 0.215
## Residuals 100 0.142798 0.0014280 0.74748
## Total 119 0.191038 1.00000
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1ANOSIM and beta dispersivity
attach(metadata)
NMDS_anosim = anosim(DESEq2.distbray, Biostimulant)
summary(NMDS_anosim)##
## Call:
## anosim(x = DESEq2.distbray, grouping = Biostimulant)
## Dissimilarity: bray
##
## ANOSIM statistic R: 0.003099
## Significance: 0.254
##
## Permutation: free
## Number of permutations: 999
##
## Upper quantiles of permutations (null model):
## 90% 95% 97.5% 99%
## 0.00778 0.01135 0.01480 0.01963
##
## Dissimilarity ranks between and within classes:
## 0% 25% 50% 75% 100% N
## Between 1 1795.50 3576.5 5360.50 7140 5760
## Control 114 2125.00 4360.0 5730.75 7139 276
## Endomaxx 771 2953.25 4398.5 6281.00 7135 276
## Inocucor 8 1840.50 3811.5 5355.00 7059 276
## Pathway 45 1690.75 3539.5 5610.25 7124 276
## Sumagrow 13 906.25 2010.0 3533.75 5491 276plot(NMDS_anosim)
beta <- betadisper(DESEq2.distbray, DESEq2.df$Biostimulant)
permutest(beta)##
## Permutation test for homogeneity of multivariate dispersions
## Permutation: free
## Number of permutations: 999
##
## Response: Distances
## Df Sum Sq Mean Sq F N.Perm Pr(>F)
## Groups 4 0.004390 0.00109760 2.6446 999 0.037 *
## Residuals 115 0.047728 0.00041503
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1CCA with VIF selection
source('https://andreanuzzo.github.io/Strausslab/vif.cca.bw_sel.R')
vifvariables = lda_variables
vifvariables = vifvariables[,which(names(vifvariables)!='LinkerPrimerSequence' &
names(vifvariables)!='BarcodeSequence' &
names(vifvariables)!='Description' &
names(vifvariables)!='InputFileName' &
names(vifvariables)!='X.SampleID' &
names(vifvariables)!='Batch' &
names(vifvariables)!='Rep')]
cca_vif <- vif.cca.bw_sel(DESEq2.phylobj, vifvariables, threshold = 5)## [1] "Removing variable Root with VIF:167.88"
## [1] "Removing variable LDA1_biomass_vs_treatment with VIF:130.9"
## [1] "Removing variable Stems with VIF:79.5"
## [1] "Removing variable TimePointT3 with VIF:25.48"
## [1] "Removing variable Leaves with VIF:22.74"
## [1] "Removing variable BiostimulantSumagrow with VIF:11.68"plot(cca_vif)
anova(cca_vif)CCA plot
cca_plot <- plot_ordination(physeq = pslog,
ordination = cca_vif,
type= 'split',
color = "Biostimulant",
label = 'Phylum'
) + aes(shape = TimePoint) +
geom_point(aes(colour = Biostimulant))
cca_arrowmat <- scores(cca_vif, display = "bp")
cca_arrowdf <- data.frame(labels = rownames(cca_arrowmat), cca_arrowmat)
cca_arrow_map <- aes(xend = CCA1,
yend = CCA2,
x = 0,
y = 0,
color = NULL,
shape = NULL)
cca_label_map <- aes(x = 1.3 * CCA1,
y = 1.3 * CCA2,
color = NULL,
label = labels,
shape = NULL)
cca_arrowhead = arrow(length = unit(0.02, "npc"))
cca_plot + geom_segment(mapping = cca_arrow_map, size = .5, data = cca_arrowdf,
color = "black", arrow = cca_arrowhead) +
geom_text(mapping = cca_label_map, size = 2, data = cca_arrowdf) +
ggtitle("CCA with Bray-Curtis distances") + scale_color_pander()## Warning: Removed 10215 rows containing missing values (geom_point).
##CAP Plot
source('https://andreanuzzo.github.io/Strausslab/vif.cap.bw_sel.R')
cap_vif <- vif.cap.bw_sel(pslog, vifvariables, threshold = 5)## [1] "Removing variable Root with VIF:167.19"
## [1] "Removing variable LDA1_biomass_vs_treatment with VIF:130.4"
## [1] "Removing variable Stems with VIF:79.59"
## [1] "Removing variable TimePointT3 with VIF:25.45"
## [1] "Removing variable Leaves with VIF:22.6"
## [1] "Removing variable BiostimulantSumagrow with VIF:11.48"cap_plot <- plot_ordination(physeq = pslog,
ordination = cap_vif,
type= 'split',
color = "Biostimulant",
label = 'Phylum',
axes = c(1,2)) +
aes(shape = TimePoint) +
geom_point(aes(colour = Biostimulant)) + scale_color_pander()
arrowmat <- scores(cap_vif, display = "bp")
arrowdf <- data.frame(labels = rownames(arrowmat), arrowmat)
arrow_map <- aes(xend = CAP1,
yend = CAP2,
x = 0,
y = 0,
color = NULL,
shape = NULL)
label_map <- aes(x = 1.3 * CAP1,
y = 1.3 * CAP2,
color = NULL,
label = labels,
shape = NULL)
arrowhead = arrow(length = unit(0.02, "npc"))
cap_plot +
geom_segment(mapping = arrow_map, size = .5, data = arrowdf,
color = "#0C3D4C", arrow = arrowhead) +
geom_text(mapping = label_map, size = 2, data = arrowdf, show.legend = FALSE) +
ggtitle("Constrained Analysis with Bray-Curtis distances") ## Warning: Removed 10215 rows containing missing values (geom_point).
Machine Learning in Qiime2
Input preparation
#!/bin/bash
# ----------------SLURM Parameters----------------
#SBATCH -J q2_model_input
#SBATCH --time=1:00:00
#SBATCH --ntasks=1
#SBATCH -D /ufrc/strauss/andrea.nuzzo/projects/Workshop/merged
#SBATCH -o logs/4_prepare_input_%j.out
#SBATCH -A strauss
date;hostname;pwd
################################################################################
#
# This script will prepare the input table for the modelling
#
################################################################################
# ----------------Housekeeping--------------------
rm -r models
mkdir models
mkdir models/input
cd models
# ----------------Load Modules--------------------
module load qiime2
# ------------------Commands----------------------
qiime taxa filter-table \
--i-table ../features/table.qza \
--i-taxonomy ../features/taxonomy.qza \
--p-exclude chloroplast,mithocondria \
--o-filtered-table input/inptab.qza
cd input
#Check the merged table
qiime feature-table summarize \
--i-table inptab.qza \
--o-visualization table.qzv \
--m-sample-metadata-file ../../../mapping_LDA.txt
#Collapse taxa levels at genus level
qiime taxa collapse \
--i-table inptab.qza \
--i-taxonomy ../../features/taxonomy.qza \
--p-level 6 \
--o-collapsed-table inptab_genus.qza
module unload qiime2
date
Gneiss with OLS and LME
#!/bin/bash
# ----------------SLURM Parameters----------------
#SBATCH -J q2_gneiss
#SBATCH --mem=20g
#SBATCH --time=21:00:00
#SBATCH --ntasks=1
#SBATCH -n 8
#SBATCH -D /ufrc/strauss/andrea.nuzzo/projects/Workshop/merged
#SBATCH -o logs/5a_q2_gneiss_%j.out
#SBATCH -A strauss
date;hostname;pwd
################################################################################
#
# This script performs Gneiss log-transformed balances, then OLS or LME modelling on your data
#
################################################################################
# ----------------Housekeeping--------------------
cd models
rm -r gneiss
mkdir gneiss
cd gneiss
# ----------------Load Modules--------------------
module load qiime2
# ------------------Commands----------------------
qiime feature-table filter-features \
--i-table ../input/inptab.qza \
--o-filtered-table filtered-table.qza \
--p-min-samples 5
qiime gneiss correlation-clustering \
--i-table filtered-table.qza \
--o-clustering hierarchy.qza
qiime gneiss ilr-hierarchical \
--i-table filtered-table.qza \
--i-tree hierarchy.qza \
--o-balances balances.qza
qiime gneiss dendrogram-heatmap \
--i-table filtered-table.qza \
--i-tree hierarchy.qza \
--m-metadata-file ../../../mappings_LDA.txt \
--m-metadata-column Biostimulant \
--p-color-map viridis \
--o-visualization heatmap.qzv
qiime gneiss ols-regression \
--p-formula "TimePoint+Biostimulant+Biostimulant_TimePoint+Area+SPAD+Leaves+Stems+Root+Tot+LDA1_area_vs_treatment+LDA1_biomass_vs_treatment" \
--i-table balances.qza \
--i-tree hierarchy.qza \
--m-metadata-file ../../../mappings_LDA.txt \
--o-visualization regression_summary.qzv
qiime gneiss lme-regression \
--p-formula "Biostimulant+Area+SPAD+Leaves+Stems+Root+Tot+LDA1_area_vs_treatment+LDA1_biomass_vs_treatment" \
--p-groups TimePoint \
--i-table balances.qza \
--i-tree hierarchy.qza \
--m-metadata-file ../../../mappings_LDA.txt \
--o-visualization lme_all_vs_time.qzv
module unload qiime2
date
Exctract Balances
Option 1: one by one, qiime2 way
#!/bin/bash
# ----------------SLURM Parameters----------------
#SBATCH -J q2_balance_tax
#SBATCH --mem=10g
#SBATCH --time=1:00:00
#SBATCH --ntasks=1
#SBATCH -n 2
#SBATCH -D /ufrc/strauss/andrea.nuzzo/projects/Workshop/merged
#SBATCH -o logs/5b_balance_taxonomy_%j.out
#SBATCH -A strauss
date;hostname;pwd
################################################################################
#
# Finally, we want to extract the taxonomy for the balances which were
# Significant in the previous steps
#
################################################################################
# ----------------Housekeeping--------------------
cd models/gneiss
rm -r y*.qzv
# ----------------Load Modules--------------------
module load qiime2
# ------------------QUESTIONS---------------------
# Before going on, try to look at the summary of the model and answer:
# How much is explained by the model? (R-squared)
# Does the model overfit? (pred_mse < model_mse)
# Do residuals show a trend? (if yes, you missed a variable)
# Are predicted in the residuals range? (Usually yes is expected for low R-squared)
# Do you have any significant balance in the OLS? (red in the heatmap)
# ------------------Commands----------------------
# If you have any significant balance and your model does not overfit, do the
# following for each balance to know who are the strains that significantly weigh
# for the chosen variable (i.e. Location)
qiime gneiss balance-taxonomy \
--i-table composition.qza \
--i-tree hierarchy.qza \
--i-taxonomy ../../features/taxonomy.qza \
--p-taxa-level 6 \
--p-balance-name y0 \
--m-metadata-file ../../../mappings_LDA.txt \
--m-metadata-column TimePoint \
--o-visualization y0_taxa_TimePoint.qzv
qiime gneiss balance-taxonomy \
--i-table composition.qza \
--i-tree hierarchy.qza \
--i-taxonomy ../../features/taxonomy.qza \
--p-taxa-level 6 \
--p-balance-name y0 \
--m-metadata-file ../../../mappings_LDA.txt \
--m-metadata-column Area \
--o-visualization y0_taxa_Area.qzv
date
Option 2: Do a batch extraction with a python script and then analize the mastodontic csv file
I wrote a small python script that can be called through a bash job. As long as you kept the folder structure consistent with the Qiime2 tutorial it will work on your files as well. The script is hosted on GitHub so you can download and modify it if needed.
#!/bin/bash
# ----------------SLURM Parameters----------------
#SBATCH -J q2_gneiss
#SBATCH --mem=20g
#SBATCH --time=21:00:00
#SBATCH --ntasks=1
#SBATCH -n 8
#SBATCH -D /ufrc/strauss/andrea.nuzzo/projects/Workshop/
#SBATCH -o merged/logs/5c_gneiss_extractor_%j.out
#SBATCH -A strauss
date;hostname;pwd
################################################################################
#
# Extraction of all balances having adjusted p-values<0.01 from the gneiss viz
#
################################################################################
module load qiime2/2018.6
wget -O merged/scripts/Balance_extractor.py https://andreanuzzo.github.io/Strausslab/Balance_extractor.py
merged/scripts/Balance_extractor.py
module purge
dateTo elaborate the mastodontic file you had to use either R again or Tableau or similar. Excel will fail poorly.
rel.abund <- transform_sample_counts(phylobj, function(x){x/sum(x)})library(data.table)##
## Attaching package: 'data.table'## The following objects are masked from 'package:dplyr':
##
## between, first, last## The following object is masked from 'package:purrr':
##
## transposeextracted_balances <- fread('merged/models/gneiss/ols_summary_dir/extracted_balances.csv')
Area.balances <-extracted_balances %>%
dplyr::select(kingdom, phylum, class, order, family, genus, species) %>%
mutate_at(vars(kingdom, phylum, class, order, family, genus, species), funs(gsub(".*__", "",.)))
subset_taxa(rel.abund, Genus %in% Area.balances$genus &
!Kingdom =='Unassigned' & !Kingdom =='Archaea' & !Kingdom =='Plantae') %>%
psmelt() %>%
group_by(TimePoint, Biostimulant, Kingdom, Phylum, Class, Area) %>%
summarize(mean=mean(Abundance, na.rm = TRUE)*100) %>%
filter(mean>1e-2) %>%
mutate(bin=ntile(Area,10)*round(min(Area, na.rm = TRUE),0)) %>%
ggplot() +
aes(x=Biostimulant, y = mean, colour=Class, size=bin) +
geom_point(position = position_jitter()) +
facet_grid(Kingdom~TimePoint) +
labs(y = "Relative Abundance %", x = "Area (sqcm)") +
theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5),
legend.box = 'vertical') +
guides(color=guide_legend(ncol=3)) +
scale_colour_discrete(guide = guide_legend(title.position = "top", nrow = 1)) +
scale_color_pander()+
ggtitle('Classes differing for Area')## Scale for 'colour' is already present. Adding another scale for
## 'colour', which will replace the existing scale.
Option 3: Jupyter notebook
The other option is to use the Qiime2 API and use a Jupyter notebook or an RStudio notebook to elaborate the balances. Requires a bit of python knowledge (you can do it!)
nano merged/scripts/jupyter_notebook.shAnd then
#!/bin/bash
#----------SLURM Parameters-----------------#
#SBATCH --job-name=jupyter_andy
#SBATCH -D /ufrc/strauss/andrea.nuzzo/Workshop/merged
#SBATCH --account=strauss
#SBATCH --output=logs/jupyter_%j.log
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=4
#SBATCH --mem=10gb
#SBATCH --time=72:00:00
date;hostname;pwd
module load qiime2/2018.6
jupyter-notebook --no-browser --port=23456 --ip='*'
datesbatch merged/scripts/jupyter_notebook.shWait a couple of minutes. You will then have to open the log. You then need to open an ssh tunnel, using the address that the log gives you. It will appear as something like The Jupyter Notebook is running at:http://(c23a-s40.ufhpc or 127.0.0.1):23456/. Copy the part before ufhpc and complete this command in a new terminal window.
ssh -NL 8000:c21b-s18.ufhpc:23456 your_email@hpg.rc.ufl.eduYou will be prompted with the request for password. After that, open a new Firefox window and digit localhost:8000 to enter. If you didn’t set your password, use the token that it’s in the jupyter notebook log you open previously.
Random Forest Analyses
#!/bin/bash
# ----------------SLURM Parameters----------------
#SBATCH -J q2_RF
#SBATCH --mem=30g
#SBATCH --time=21:00:00
#SBATCH --ntasks=1
#SBATCH -n 8
#SBATCH -D /ufrc/strauss/andrea.nuzzo/projects/Workshop/merged
#SBATCH -o logs/6_q2_RF_%j.out
#SBATCH -A strauss
date;hostname;pwd
################################################################################
#
# This script performs Regression/classification with RF on your data
#
################################################################################
# ----------------Housekeeping--------------------
cd models
rm -r randomforest
mkdir randomforest
cd randomforest
# ----------------Load Modules--------------------
module load qiime2/2018.6
# ------------------Commands----------------------
#Classification on categorigal variables
qiime sample-classifier classify-samples \
--i-table ../input/inptab_genus.qza \
--m-metadata-file ../../../mappings_LDA.txt \
--m-metadata-column TimePoint \
--p-optimize-feature-selection \
--p-parameter-tuning \
--p-estimator RandomForestClassifier \
--p-n-estimators 500 \
--p-cv 5 \
--p-random-state 42 \
--p-n-jobs -1 \
--p-palette GreenBlue \
--output-dir RFC
qiime sample-classifier regress-samples \
--i-table ../input/inptab_genus.qza \
--m-metadata-file ../../../mappings_LDA.txt \
--m-metadata-column Area \
--p-optimize-feature-selection \
--p-parameter-tuning \
--p-estimator RandomForestRegressor \
--p-n-estimators 500 \
--p-cv 5 \
--p-random-state 42 \
--p-n-jobs -1 \
--output-dir RFR
module unload qiime2
date
Elaboration of random forest results
rel.abund <- transform_sample_counts(phylobj, function(x){x/sum(x)})#These are the feature importances >1% in the RandomForest analysis made on qiime2
time.importance <-read.delim('merged/models/randomforest/Timepoint_feature_importance.tsv') %>%
separate(feature, c("Kingdom", "Phylum", "Class", "Order", "Family","Genus"), sep=';') %>%
mutate_at(vars(Kingdom, Phylum, Class, Order, Family, Genus), funs(gsub(".*__", "",.))) %>%
filter(importance>1e-2)
subset_taxa(rel.abund, Genus %in% time.importance$Genus &
!Kingdom =='Unassigned' & !Kingdom =='Archaea') %>%
psmelt() %>%
group_by(TimePoint, Biostimulant, Kingdom, Genus) %>%
summarize(mean=mean(Abundance, na.rm = TRUE)*100,
stdev=sd(Abundance, na.rm = TRUE)) %>%
ggplot() +
aes(x=TimePoint, y = mean, fill=Genus) +
geom_bar(stat = 'identity') +
scale_fill_pander() +
facet_grid(Kingdom~Biostimulant) +
theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5)) +
labs(y = "Relative Abundance %", x = "Filtered Samples") +
ggtitle('Species differing over time')
Area_RF <- read.delim('merged/models/randomforest/Area_feature_importance.tsv') %>%
separate(feature, c("Kingdom", "Phylum", "Class", "Order", "Family","Genus"), sep=';') %>%
mutate_at(vars(Kingdom, Phylum, Class, Order, Family, Genus), funs(gsub(".*__", "",.))) %>%
filter(importance>1e-2)
subset_taxa(rel.abund, Genus %in% Area_RF$Genus &
!Kingdom =='Unassigned' & !Kingdom =='Archaea') %>%
psmelt() %>%
group_by(TimePoint, Area, Biostimulant, Kingdom, Genus) %>%
summarize(mean=mean(Abundance, na.rm = TRUE),
stdev=sd(Abundance, na.rm = TRUE),
Area.mean = mean(Area, na.rm = TRUE),
Area.stdev = sd(Area, na.rm = TRUE)) %>%
write.csv('merged/models/randomforest/Area_RF.csv')